Siddhant Bharadwaj

I am currently a PreDoc at the Indian Institute of Science, Bangalore where I work on visual quality assessment for AI generated images using MLLMs under Prof. Rajiv Soundararajan, and on the AI side of the Oral Cancer Screening Project under Prof. Rajesh Sundaresan and Prof. Chandra Sekhar Seelamantula.

In addition to my work at IISc, I am collaborating with Prof. Min Xu at Carnegie Mellon University on adapting 2D image-pretrained Transformers to 3D Cryo-ET subtomogram classification, and with Prof. Shruti Vyas at University of Central Florida on MLLM-based geolocalization. I also lead Computer Vision efforts at CORD.ai, a non-profit deep learning research community with 800+ members across 12 countries, and contribute to Cohere for AI's Maya project on spatial reasoning in VLMs.

I graduated with a major in Electrical and Electronics and a minor in Data Science at Manipal Institute of Technology, Manipal in 2024. My interest lies in deep learning, computer vision, and image processing. My current research primarily focuses on Visual Language Models (VLMs) and Multimodal Large Language Models (MLLMs).

Previously, I interned at Spectrum Lab in the Indian Institute of Science, Bangalore, where I worked on adapting the Segment Anything Model for the task of optic disc and optic cup segmentation in fundus images.

During my undergraduate, I conducted research on AI in Health Care, AI Security and the use of Deep Learning in battery health management. The bulk of the work was done under Prof. Harish Kumar J.R. (MIT, Manipal), and Prof. Munesh Chandra Trivedi (NIT Agartala).

When I am not programming or doing mathematics, you will find me reading, watching sitcoms, scribbling in my journal, and more often than not posting hot takes on X (formerly known as Twitter).

If you would like to collaborate or talk about one of my projects, feel free to drop a mail.